ALONE: A Dataset for Toxic Behavior among Adolescents on Twitter

TL;DR

This paper introduces ALONE, a multimodal dataset of adolescent Twitter interactions, highlighting the importance of context and multiple modalities in detecting toxic behavior among teenagers.

Contribution

The paper presents a novel multimodal dataset with tweets, images, emojis, and metadata, specifically focused on adolescent interactions to improve toxicity detection.

Findings

Context-aware analysis improves toxicity detection accuracy.

Individual tweets alone are insufficient for identifying toxicity.

Multimodal data enhances understanding of toxic interactions.

Abstract

The convenience of social media has also enabled its misuse, potentially resulting in toxic behavior. Nearly 66% of internet users have observed online harassment, and 41% claim personal experience, with 18% facing severe forms of online harassment. This toxic communication has a significant impact on the well-being of young individuals, affecting mental health and, in some cases, resulting in suicide. These communications exhibit complex linguistic and contextual characteristics, making recognition of such narratives challenging. In this paper, we provide a multimodal dataset of toxic social media interactions between confirmed high school students, called ALONE (AdoLescents ON twittEr), along with descriptive explanation. Each instance of interaction includes tweets, images, emoji and related metadata. Our observations show that individual tweets do not provide sufficient evidence for toxic behavior, and meaningful use of context in interactions can enable highlighting or exonerating tweets with purported toxicity.